This application responds to PAR-14-008 for secondary analyses of existing clinical research datasets in order to assess the validity of the [Research Domain Criteria] RDoC constructs, and to test novel hypotheses using the RDoC framework. RDoC was introduced in 2009 as a new classification framework for mental disorders. It is based on a matrix of 5 domains (systems) of functioning, each of which includes lower order constructs, and each construct is defined across multiple units of analyses (UAs) (e.g., neural circuits, genes, physiology, actual behavior, and self-report). Given that RDoC is still a moving target (Cuthbert & Insel, 2013), we propose to help refine it by examining the construct and predictive validity of selected constituents. Our archival data derive from a 10-year Program Project on risk factors for juvenile- onset depression (P01-MH-056193), which included (a) adults and juveniles, ranging from minimally symptomatic controls to those with multiple co-morbid psychiatric syndromes and (b) studies of emotional reactivity and regulation in the context of various gradations of familial and personal risk for depression. Thus, the data are well suited to examine facets of RDoC's Negative/Positive Valence Systems. Relevant to the Negative Valence System, we have experimental analogues probing physiological, neural, and psychological responses to Loss, Frustrative Non-reward, and Potential Threat. Relevant to the Positive Valence System, we have probes of Initial Response to Positive Reinforcement and Reward Expectancy. Our neural index is frontal lateralization of EEG alpha waves (EEG asymmetry); one physiological index is respiratory sinus arrhythmia (RSA), an indicator of autonomic nervous system functioning related to affect experience and adaptive functioning. Our behavioral indexes include self-, clinician-, and observer-based ratings of affect, symptoms, and functioning. We have follow-up assessments of key Constructs and outcomes. We propose two Aims. Specific Aim 1. Examine the construct validity of the Negative/Positive Valence Systems in adults and juveniles using neural, physiological, and behavioral UAs. This Aim will be guided by 4 Hypotheses (that include developmental and familial analyses) concerning the relations of UAs and Constructs within and across the Positive and Negative Valence Systems. Specific Aim 2. Examine the predictive validity of Negative and Positive Valence System Constructs in relation to distress (depression/anxiety symptoms), trait negative and positive affect, and adaptive functioning. This Aim will be guided by 2 Hypotheses concerning the concurrent and longitudinal predictive power of Positive/Negative Valence systems and related UAs and constructs. While implementing standard psychometric approaches to validity testing, our study is unique owing to its longitudinal aspect and its developmental & familial emphases: it includes child samples and proposes that Constructs run in the family. Our findings will help pave the way for next steps in the refinement of the RDoC matrix.